端口速率

典型的ATM输出缓冲交换结构有着理想的性能，但需要N倍端口速率的高速传送而难以实现。

The typical model of output queueing in ATM switching has ideal performance , but is difficult to realize because of the need of high speed transfer that is N times the port speed .

一种端口速率可变的共享存储器交换

A variable port speed shared memory switch

输入缓存交换结构的特点是缓存器和交换结构的运行速率与端口速率相等、实现容易，但存在队头阻塞（HOL），其吞吐率只有约58%。

Input queued cell switch has the advantage of same bandwidth of buffer and switching architecture as that of inputs and outputs . But the head of line blocking in input queued switch limits its throughput to approximately 58 % .

但是对于路由器而言，一般路由器的瓶颈在于转发引擎的处理速率而不在于端口速率。

The bottleneck of the router is not depended on the speed of the port , but depended on by the speed of the forwarding engine .

在输入端、输出端同时设立缓冲队列（CIOQ）的交换结构可以完全模拟典型输出缓冲交换结构，而且仅需2倍端口速率的传送。

A model that sets buffers at both input and output ports can simulate the typical model of output queueing at a transfer rate only twice the port speed .

一次查找仅需3步搜索、转发表无需排序、表项更新快、查找速率高且连续性好，满足IPv4/IPv6双栈核心路由器OC-768（40Gbit/s）端口的线速率转发。

Its lookup speed is high , entry updating completes fast , and lookup continuity was enhanced . It achieves OC-768 ( 40Gbit / s ) ports ' wire-speed forwarding and satisfies IPv4 / IPv6 dual-stack core routers .

其次，在网络处理器平台上设计了瓶颈带宽测量软件，可用于多网网关的端到端的瓶颈链路带宽测量，论文还设计了多网网关的端口输入输出速率的测量方案。

Secondly , this dissertation designs the bottleneck bandwidth measurement software for the platform based on the network processor , which is used to measure end-to-end bottleneck bandwidth . In addition , the measurement scheme of the input - and output - port rates for the multi-network gateway is designed .